Crypto market cycles are the multi-year boom-and-bust sequences that define bitcoin's price history: an accelerating advance, a blow-off peak, a decline that has historically run between about 77 and 93 percent, then a long basing period before the next advance. Bitcoin has traced four such arcs since 2011, with peaks landing close to four years apart — hence the four-year cycle. The popular explanation ties that rhythm to the halving, the programmed cut in new bitcoin issuance that occurs every 210,000 blocks.
The short answer for anyone searching this topic: the halving is a real supply event, but its usefulness as a market timer has weakened with each repetition, and the historical cycles line up at least as well with global liquidity and interest-rate regimes as with bitcoin's issuance schedule. The rest of this page works through why.
The halving cadence
At about ten minutes per block, 210,000 blocks take just under four years to mine. The reward fell from 50 coins per block to 25 in November 2012, to 12.5 in July 2016, to 6.25 in May 2020, and to 3.125 in April 2024. At the current subsidy and around 144 blocks per day, miners receive about 450 new coins daily.
The bull case is simple: demand held constant, a scheduled supply cut should push price up. Because each historical halving preceded a major advance by several months to a year, the schedule acquired a reputation as a clock.
Why the clock keeps losing accuracy
Four separate forces erode it:
- Flow shrinks relative to stock. Around the 2012 halving, new issuance ran above 20 percent of outstanding supply on an annualized basis. Today it sits under one percent. Each halving cuts a smaller absolute flow, while daily traded volume across spot and derivatives dwarfs miner issuance many times over.
- Anticipation. The halving date is public years in advance, and a widely believed, perfectly scheduled catalyst invites front-running — the more famous the pattern, the earlier participants position for it.
- Sample size. Four halvings is four observations. A cycle length estimated from n = 4 carries error bars wide enough to drive a year through. Each successive advance has also been smaller in multiple terms, which cuts both ways: diminishing returns are plausible, but so is coincidence.
- Confounding events. The May 2020 halving landed weeks after the largest emergency monetary expansion in modern history. Attributing the 2020–2021 advance to the issuance schedule means ignoring the macro backdrop entirely.
Liquidity and rate regimes underneath the pattern
A cleaner explanation carries an actual mechanism. Crypto behaves like a long-duration risk asset with no cash flows, acutely sensitive to real interest rates and the supply of speculative capital. When policy rates sit near zero and central-bank balance sheets expand, the opportunity cost of holding a volatile non-yielding asset collapses. When rates rise quickly, the same asset gets repriced hard.
The historical record fits. The 2020–2021 advance unfolded during near-zero rates plus unprecedented fiscal and monetary expansion. The 2022 decline tracked the fastest hiking cycle in four decades, a stretch in which bitcoin traded in tight correlation with long-duration growth equities. Global money-supply measures have lined up with crypto's major turns, though the lead-lag is unstable enough to make the correlation context rather than trigger.
A distinction worth keeping: macro liquidity is not order-book liquidity. The depth visible on a depth-of-market ladder is a microstructure quantity, and crypto books thin out during off-hours and stress. The two interact — a macro shock hitting a thin book produces the liquidation cascades that punctuate every cycle — but they are separate variables.
Trading the regime fails in its own way: policy inflections are obvious only in hindsight, so waiting for the pivot means buying after much of the repricing, while anticipating it means eating repeated small losses through every false transition.
Drawdown history as the honest risk frame
Approximate peak-to-trough declines for bitcoin, cycle by cycle: about 93 percent in 2011, about 85 percent from the 2013 peak into 2015, about 84 percent from late 2017 into December 2018, and about 77 percent from November 2021 into late 2022. The declines have grown shallower, but not one has stopped short of 75 percent. Most large altcoins fell further in each episode, and the majority never returned to their prior highs.
For anyone trading this asset class with leverage, that record is the binding constraint. One explicitly hypothetical example: suppose a cash-settled futures contract carries a 5-coin multiplier and trades at a hypothetical $20,000, for $100,000 of notional against $40,000 of posted margin — 2.5x effective leverage. A 40 percent decline to $12,000 produces a loss of $8,000 × 5 = $40,000: the entire margin, on a move about half the size of the mildest cycle drawdown on record.
Founder note: I sat through the 2021–2022 decline. What stuck was not the depth of the move but its duration — an account survives a crash more easily than it survives a year of lower highs grinding against margin.
The frame gets misused in two directions. Sizing to the worst historical decline assumes the sample already contains the worst outcome; 77 to 93 percent is an observed range, not a floor, and the next cycle owes nothing to it. Treating a fixed drawdown as a value signal — down 70 percent means cheap — loses on any asset headed to zero, which is the modal outcome for small-cap tokens across every cycle so far.
On-chain versus price-derived cycle measures
Cycle position gets measured in two broad families that answer different questions.
On-chain measures come from the blockchain itself: realized price (the aggregate cost basis of all coins when last moved), MVRV (market value over realized value), spent-output profit ratios, long-term-holder supply, exchange balances. Their strength is information the tape does not contain — what holders actually paid, and how long coins have sat dormant. Their weakness is structural fragility. Custodial consolidation and exchange internal transfers blur entity boundaries, and spot ETF custody concentrated enormous holdings behind a handful of address patterns. An "entity-adjusted" metric is an estimate wearing the costume of a measurement.
Price-derived measures need only the chart: multi-year moving averages, ratios of price to a long average, drawdown from the prior high, realized-volatility regimes. Their strength is that nothing is estimated. Their weakness is lag by construction — a 200-week average confirms a cycle turn long after it happened, on any asset, every time.
Intraday work has the same split: traded-flow measures such as cumulative delta describe what transacted; the resting book describes what might. The cycle versions inherit the same discipline: on-chain data says who owns coins at what basis; price-derived data says what the trend has already done.
Valuation-band strategies built on either family — accumulate below realized price, reduce at historically stretched MVRV — lose when a structural break shifts the band, because a metric can read cheap far longer than a leveraged position can be financed. Any band calibrated on past cycles also inherits the small-sample problem above.
Survivorship bias in the cycle charts
The overlay chart that launched a thousand theories — four cycles aligned from their bottoms, each recovering to a new high — is a survivor's autobiography. It exists because bitcoin recovered every time. No such chart can be drawn for most of the top-100 assets of 2013 or 2017, which never saw their prior peaks again. The four-year pattern is a property of the one asset that lived, presented as if it were a property of the asset class.
Construction choices add their own distortion. Log scaling compresses 90 percent declines into visually gentle dips. Aligning cycles from their bottoms smuggles hindsight into the x-axis, since bottoms are only identifiable years later. Narrative selection then operates on theories the way markets operate on tokens — flow-based scarcity models that published precise price paths for 2021 were falsified and quietly retired, while the surviving story keeps getting retold. Backtesting a cycle strategy on bitcoin alone therefore conditions on survival twice: once in the asset, once in the theory.
Using the cycle as context, not a countdown
Where this leaves a futures or options trader: the cycle frame earns its place as context for position sizing and exposure duration, not as an entry signal. History says spot declines beyond 75 percent are normal behavior for this market, so effective leverage has to be set as if one can begin at any time. The macro-liquidity question and the timing question live on different clocks — the first moves over quarters, the second on execution timeframes — and confusing them produces positions with a thesis measured in years and margin measured in days. Transition zones are where every framework here fails at once: each measure above is a slow variable, and slow variables read best precisely when acting on them has become expensive. The four-year chart is a hypothesis resting on four data points. Price the risk as if it can break, because nothing in its construction says it cannot.